Growth was studied in 68 children aged 2-12 years with atopic eczema. Height SD scores were significantly correlated with the surface area of skin affected by eczema. The mean height of41 patients with less than 50/o of their skin surface affected (group I) was normal (mean SD score -0.11). The 27 children with more than 50% of their skin affected (group II) were sigiicantly shorter (SD score -0.83) and were also short allowing for their parental target height. The predicted heights were also normal in group I but were lower than expected in group H. Regression analysis suggested that height was most dependent on parental height. The extent of the disease had a significant additional effect, whereas dietary treatment and treatment with topical steroids had only marginal additional effects. The growth of children with eczema affecting less than 50% of the skin surface area appears to be normal, and impaired growth is confined to those with more extensive disease.

It has been suggested that delirium in the elderly is caused by abnormally high levels of circulating glucocorticoids or by an increased vulnerability to their effects. We performed a dexamethasone suppression test (DST) in 16 consecutive patients without depression or dementia admitted to an acute-care geriatric unit with a clinical diagnosis of lower respiratory tract infection. Seven of 9 (78%) patients who developed delirium were non-suppressors on the DST compared with 1 of 7 (14%) patients without delirium (p = 0.04). Clinical and laboratory indicators of the severity of illness did not differ between the two groups. Of the 8 patients with an abnormal DST, 1 died and another was not available for repeat assessment. On re-examination 8 weeks later, after resolution of the delirium and of the chest infection, 5 of 6 non-suppressors still had an abnormal DST. It is known that some non-demented and non-depressed elderly patients fail to suppress cortisol in response to 1 mg of dexamethasone. Our results suggest that such patients may be at increased risk for developing delirium during acute illness.

The hypothesis that atopic eczema is associated with a non-specific decrease in the serum concentration of trace metals was examined by measurement of the levels of zinc, copper and iron in 134 children with atopic eczema and 112 controls. The results failed to confirm the hypothesis. There was no significant difference between patients and controls for the concentrations of serum zinc, iron, iron-binding capacity and albumin. Patients with eczema had a significantly higher copper concentration (median 21.0 mumol/l) than controls (median 17.0 mumol/l), and a significantly lower concentration of ferritin (median 11.9 ng/ml for patients and 16.5 ng/ml for controls). There was a highly significant correlation between the surface area of skin affected by eczema and the concentrations of albumin and orosomucoid.

Four children with severe atopic eczema developed painless endochondral pseudocysts of the external auricle. The lesions were bilateral in two cases. In one case the lesion spontaneously resolved; of the remaining five lesions, one contained haematoma and the other four contained serous fluid. The aetiology of these lesions and their association with atopic eczema are unclear, but repeated minor trauma from rubbing may play a part.

In children with atopic eczema on elimination diets, the calcium intake was below the estimated requirement in 15 out of 20 who avoided cows' milk and received no milk substitute, and in three out of 26 who avoided cows' milk but were provided with a soya or casein hydrolysate formula.